A multilevel data reuse scheme for Motion Estimation and its VLSI design

Abstract

Motion Estimation (ME) in video coding is a vital component that excels not only in computational complexity, but off-chip memory bandwidth as well. These two issues are considered critical constraints in terms of High Definition (HD) video coding, since a large volume of data must be processed. The multilevel data reuse scheme proposed in this paper is able to reduce the off-chip memory bandwidth, with direct impact in throughput and energy consumption. This scheme explores the concept of overlapped Search Windows (SW) in more than one level and poses no harm to video quality. Comparisons with related works show that this solution provides the best tradeoff between the use of on-chip memory and reduction of the off-chip memory bandwidth. The data reuse scheme was applied in a ME architecture and the synthesis results show that this solution presented the lowest use of hardware resources and the highest operation frequency among related works. The proposed architecture is able to process 1080p videos at 25 fps, and the reduction ratio of off-chip memory access achieved by the architecture is greater than 95% when compared to the traditional method.

Memory energy consumption reduction in video coding systems

“…Work [5] presents a multilevel data reuse scheme based on overlapped search windows among neighboring blocks. The proposed scheme can greatly reduce the number of off-chip and on-chip memory accesses with no penalty on video quality.…”

“…When real time applications are considered, a high level of parallelism must be used, exploiting the inherent parallelism of FS to achieve the required performance. However, the use of high level of parallelism requires also a high memory bandwidth increasing the hardware consumption and also energy consumption as well [5].…”

“…Solutions to reduce external memory accesses are based on two main approaches: (1) external memory bandwidth reduction through data reuse strategies using caches [5], [6] and [7], and (2) bandwidth reduction by reference frames compression before they are stored in the external memory [4], [8], [9] and [10]. This paper presents a memory consumption reduction scheme that uses these two approaches, external memory bandwidth reduction through data reuse schemes [5] and bandwidth reduction through reference frames compression [4].…”

“…This paper presents a memory consumption reduction scheme that uses these two approaches, external memory bandwidth reduction through data reuse schemes [5] and bandwidth reduction through reference frames compression [4]. These two techniques are combined to achieve an average compression rate of 86.32% for reading and 31.69% for writing operations over reference frames in a video coding system.…”

An Optimized Hardware Video Encoder for AVS with Level C+ Data Reuse Scheme for Motion Estimation

“…In paper [3], Level C+ data reuse scheme and corresponding zigzag coding order were proposed to reduce the required external memory bandwidth. In [4] [5], Level C+ scheme has been implemented into the encoder to reduce bandwidth but these papers do not give enough details on the Level C+ implementation. Furthermore, this paper will mainly present the special considerations for the implementation of a Level C+ encoder that the traditional Level C encoder does not need to care.…”